Understanding How Psychological Distance Influences User Preferences in Conversational Versus Web Search

TL;DR

This paper investigates how psychological distance, across spatial, temporal, social, and hypothetical dimensions, shapes user preferences between web search and LLM-powered conversational search (ConvSearch) in a travel-planning task. Using a 2x4 between-subjects design with 128 participants, it demonstrates that greater distance amplifies ConvSearch advantages in ease of use, credibility, usefulness, enjoyment, and overall acceptance, while near-distance effects are weaker. The authors integrate Construal Level Theory with TAM-based measures and conduct thematic analyses to reveal qualitative reasons behind system preferences, identifying Low Workload, Speed and Efficiency, and Interactive dialogue as ConvSearch strengths, versus Visual Confirmation and Source Credibility for WebSearch. Practically, the findings support distance-aware IR design and hybrid search architectures that adapt information level and presentation to the user’s cognitive state, enhancing user experience and satisfaction in increasingly complex information environments.

Abstract

Conversational search offers an easier and faster alternative to conventional web search, while having downsides like lack of source verification. Research has examined performance disparities between these two systems in different settings. However, little work has considered the effects of variations within a given search task. We hypothesize that psychological distance - one's perceived closeness to a target event - affects information needs in search tasks, and investigate the corresponding effects on user preferences between web and conversational search systems. We find that with greater psychological distances, users perceive conversational search as more credible, useful, enjoyable, and easy to use, and demonstrate increased preference for this system. We reveal qualitative reasons for these differences and provide design implications for search system designers.

Figures (7)

• Figure 1: Extended TAMs incorporating perceived enjoyment and information credibility. This model illustrates the relationships between perceived ease of use, perceived usefulness, perceived enjoyment, and information credibility with the acceptance intention of technology.
• Figure 2: Flowchart of the study procedure detailing the sequence of tasks and surveys conducted across two information seeking tasks. This diagram illustrates the comprehensive steps from the initial pre-task survey through two distinct information seeking tasks, followed by post-task surveys and a final preference survey.
• Figure 3: Comparison of perceived ease of use between WebSearch and ConvSearch in Far Psychological Distances. Panels (a)-(d) show results at a near psychological distance, and panels (e)-(h) at a far psychological distance, highlighting significant differences in the latter. Significance is marked as p > 0.05 (ns), p <= 0.05 (*), p < 0.01 (**), or p < 0.001 (***).
• Figure 4: Impact of psychological distance on perceived information credibility for WebSearch and ConvSearch. Panels (a)-(d) illustrate results at a near psychological distance, with significant differences only in spatial distance. Panels (e)-(h) demonstrate increased credibility for ConvSearch at a far psychological distance across all dimensions except temporal distance. Significance is marked as p > 0.05 (ns), p <= 0.05 (*), p < 0.01 (**), or p < 0.001 (***).
• Figure 5: Comparative analysis of perceived usefulness between WebSearch and ConvSearch at varying psychological distances. Panels (a)-(d) show minimal differences at a near psychological distance, except for social distance. Panels (e)-(h) depict significant increases in perceived usefulness for ConvSearch at a far psychological distance across all dimensions. Significance is marked as p > 0.05 (ns), p <= 0.05 (*), p < 0.01 (**), or p < 0.001 (***).